Skip to main content
GET
Python

Authorizations

Authorization
string
header
required

API Key authentication. Format: "Bearer YOUR_API_KEY"

Path Parameters

failure_id
string<uuid>
required

Unique identifier for the failure

Query Parameters

include
string

Comma-separated extras. context adds the knowledge and rubric content that GET /api/dataframer/context/ serves, so one call gets both

exclude
string

Comma-separated lists to leave out of the response, keeping their counts and averages: reviews, judgments. Both are served by default here, unlike on the list

Response

The failure with its root-cause analysis

id
string<uuid>
name
string

Short name for the failure

description
string

What the traces in this failure have in common

project
object

The project the traces came from

trace_count
integer

Traces currently assigned to this failure

first_tracked_at
string<date-time>

When the failure was first tracked

last_refreshed_at
string<date-time> | null

When traces were last pulled and matched (null if never)

rubrics
object[]

Rubrics graded against on this failure's traces (omitted when detail=summary)

human_review
object | null

Expert-review scores (null if none; omitted when detail=summary)

llm_judge
object | null

LLM-judge scores (null if none; omitted when detail=summary)

analysis
object

Automatic root-cause analysis

content_location
object | null

Where description is stored in your own cloud (null unless your deployment is customer-hosted)

content_unavailable_reason
enum<string> | null

Set when description is empty because it is stored in your own cloud, not because there is none (null otherwise). The name, counts and scores are still complete. The Python SDK reads the text from your data plane, fills the fields and clears this

Available options:
result_content_in_data_plane
context
object

Knowledge and rubric content; returned only when include=context